[governance] IPv4 - IPv6 incompatiblity (was Re: Towards Singapore)

[Avri Doria]

Hi,

A few quibbles:


On 17 Jun 2011, at 14:44, Paul Wilson wrote:

> 
> 
> Yes, at one level v4 and v6 are incompatible, naturally, because they are Different Protocols; but above and below the IP protocol level, they are perfectly "compatible" - they run side by side on the same wires, through the same equipment, and on the same services as each other; and they support the same applications, which work in the same way, to the extent that users don't even need to know.

Well as long as the applications have been reworked to use the updated APIs that use both.  sure. Getting people to buy new applications is always useful for profits.

> 
> I've promoted an analogy between the v4-v6 transition and the transition from oil to electricity in our transport system.  And it works for this discussion as well:  you don't try to plug your volkswagon beetle into the mains - because oil and electrons certainly don't mix - but that old car is still perfectly "compatible" with the latest electric one: it drives on the same roads, uses the same rules and the same controls; and carries the same passengers in the same way.

At a very gross level, sure.  But just try to reuse the engine parts from in the other.  Not sure this analogy is that useful to making your case.  But it is good for the vendors to promote incompatibility.

> 
> As for "backward compatibility" I suggest to be careful what you ask for here, because that is always a temporary benefit, and often a long-term curse. MS Windows users have suffered vast costs and complexities for many years, just so that a few MS-DOS applications could keep running; and then there's the old  QWERTY keyboard.

> On the other hand, we might remember the complaints surrounding Apple's change to OSX, a completely new and incompatible operating system (well, being based on Unix, a completely OLD operating system).  But does anyone care about that any more?  No, it's been properly forgotten, just as IPv4 will be when the big transition is done.

Yeah, but OSX offered a lot more for the pain of the switch than IPv6 does.  To the happy IPv4 user, IPv6 offers nothing special.  Except maybe for the risk one will go through will all the bugs get worked out.

> 
> I admit to blissful ignorance of the blow-by-blow disputes and politics of the development of IPNG in its various early flavours; and I don't care much to go back there.  But I can certainly imagine that if IPv6 were shackled with tricks to have it interconnect directly with IPv4, at the IP level, then in a few years time, and for decades afterwards, we'd all be cursing the developers for their shortsightedness.

Well I have still never heard reasoning that made any sense to me about why coexistence of the addressing formats was not worked on other that the hubris of the IPv6 proponents believing that IPv6 should replace IPv4.  I think the need to maintain two routing setups is as wasteful as it is profitable for the routing vendors.

GSE, 8+8, variable length addressing or other proposed solutions for address compatibility would have gone a long way to lessening the problems of incompatible addressing.  Even after all these years, I do not understand any reason why these types of solutions did not prevail to make the networks coexist.  All I can see as reasons are profit for the hardware vendors and control of the address space by the RIR (IPv6 is controlled completely, while IPv4 is still somewhat outside their complete control).  That and the hubris of 'transition'.

And I think this decision is one of the things that makes universal deployment of IPv6 such a challenge.  Had people allowed for a smooth coexistence, perhaps IPv6 would already be deployed and IPv4 would already be withering away without anyone noticing.

> 
> My view is that the only significant sin of the IPv6 developers, at least the only one which is relevant here, is to have underestimated the coming success of the Internet.  It is that success which allowed the Internet to grow so vast and to become so very cheap, the major factors which conspire to make the IPv6 transition much harder than anyone thought.


I think the problem is that we still have people thinking of transition and not coexistence.  And I think it is this that will create problems for developing economies.  To bootstrap themselves onto the Internet, many rely on older and used technology. To then require them to replace to  a well tested networking technology  for one that still have growing pains remains a big and expensive problem.

a.


------
Pick your poison: Kool-Aid or Hemlock!




____________________________________________________________
You received this message as a subscriber on the list:
     governance at lists.cpsr.org
To be removed from the list, visit:
     http://www.igcaucus.org/unsubscribing

For all other list information and functions, see:
     http://lists.cpsr.org/lists/info/governance
To edit your profile and to find the IGC's charter, see:
     http://www.igcaucus.org/

Translate this email: http://translate.google.com/translate_t